University of Pretoria etd
CHAPTER ONE
INTRODUCTION
1.1
OBJECTIVES
Commercialisation of subsistence agriculture implies increased participation,
or, rather, an improved ability to participate, in output markets.
In the
developing areas of South Africa, like in other developing countries,
smallholder farmers find it difficult to participate in markets because of a range
of constraints and barriers reducing the incentives for participation. These
may be reflected in hidden costs that make access to markets and productive
assets difficult.
Transaction costs, that is, observable and non-observable costs associated
with exchange, are the embodiment of access barriers to market participation
by resource poor smallholders (Coase, 1960; Delgado, 1999; Holloway et al,
2000). These include the costs of searching for a trading partner with whom to
exchange, the costs of screening partners, of bargaining, monitoring,
enforcement and, eventually, transferring the product to its destination (Jaffee
and Morton, 1995; Hobbs, 1997). Transaction costs, however, do not only
include the costs of the exchange itself, but also encompass costs associated
with the reorganisation of household labour and other resources in order to
produce enough for the market.
The main objective of this study is to investigate the extent to which
transaction costs affect the market participation behaviour of smallholder
farmers in the Northern Province of South Africa. The identification of these
transaction cost factors could assist in identifying policy interventions and/or
institutional innovations to alleviate constraints and improve the ability of
small-scale farmers to be part of the commercial agricultural economy.
Transaction costs differ between various households due to asymmetries in
1
University of Pretoria etd
access to assets, information, services and remunerative markets. The study
will therefore also investigate the factors contributing to different levels of
transaction costs amongst households.
The specific objectives of the study are:
•
to identify underlying transaction cost factors influencing household
behaviour in market participation in the Northern Province of South Africa,
•
to identify factors that influence the decision of these farmers to participate
in output markets,
•
to identify factors that could contribute to increased participation in
agricultural output markets, and
•
to make recommendations to support policy formation and implementation
of agricultural development programmes.
Thus, this study aims to find ways of accelerating the participation of
previously disadvantaged farmers into mainstream agriculture. It attempts to
assess the extent to which institutional factors, particularly transaction costs,
are responsible for a low participation rate, and attempts to explain this. The
study also aims at suggesting ways to create an appropriate environment for
emerging farmers to contribute to rural growth.
1.2
JUSTIFICATION
South Africa continues to strive for empowerment of those who were denied
opportunities under apartheid. The process of empowerment is about giving
disadvantaged communities and individuals more choices, and, in the case of
agriculture, removing the dualism and fully integrating and democratising the
sector (Kirsten, van Zyl and Vink, 1998). This process is important both for
sustainable economic growth and for the alleviation of poverty and inequality.
Various efforts to promote small-scale farming have been noted in the past
decade. It remains evident, however, that much more needs to be done to
make a positive difference in terms of the political objective of an integrated
2
University of Pretoria etd
agricultural sector.
Integration will only happen when smallholder farmers
fully participate in the market.
Farmers in the neglected and less developed rural areas are generally poor.
According to a discussion paper on food security (DALA, 1997; MALA, 1998),
many households are vulnerable to food insecurity. Unemployment is high
and tends to rise as household members lose jobs in the urban centres.
Farmers in these areas are not really part of commercial agriculture. This is
one of the reasons that the contribution of smallholder agriculture to the gross
national product is still limited in South Africa. The majority of disadvantaged
farmers is not part of mainstream agriculture and practises subsistence
agriculture in overcrowded, semi-arid areas in the former homelands. This
kind of subsistence farming is characterised by low production (and
productivity), poor access to land, and poor access to inputs and credit. In
order to generate enough income these farmers tend to engage in off-farm (or
non-farm) income generating activities.
It is, however, possible for smallholder farming to survive economically when
given a set of opportunities. After all, subsistence farmers are used to take
rational decisions in order to adapt to conditions they find themselves in. For
example, given a set of resources, farmers will strive to optimise production.
Another particular and critical set of opportunities involves opening access for
smallholders to interact with other economic agents.
To a large extent the process of agricultural transformation in South Africa
involves moving households from subsistence production to producing for the
market.
Producing for the market provides a number of benefits and
advantages. In particular rural employment is promoted and income is
generated (Ngqangweni, 2000). The commercialising environment provides a
potential for increased production and thus for improving food security for the
rural poor.
Furthermore, several studies (Ngqangweni, 2000; Delgado,
Hopkins, Kelly et al, 1998) have shown positive and strong multiplier effects of
investing in agriculture. In other words, agriculture has an important role to
play in fostering rural development and poverty alleviation.
It is through
3
University of Pretoria etd
commercialisation
of
smallholder
agriculture
that
the
previously
disadvantaged groups can become a significant part of the economic base of
rural economies.
Very few smallholder farmers participate in the markets.
A range of
impediments for market participation has been identified. They include lack of
assets, market information and training. An added factor is that farmers are
located far away from the market and have poor access to infrastructure.
Several studies (Van Rooyen, Vink and Christodoulou, 1987; Kirsten et al,
1993; and Kirsten, 1994) have in the past referred to the need for structural
reform if participation of black farmers in the commercial agricultural sector is
to be enhanced.
It is to be appreciated that efforts to promote structural change, such as land
reform, improved access to credit and a number of markets, have benefited
some, albeit a small minority of black farmers. But these reforms have not
been sufficient to improve the participation in commercial agriculture of the
majority of subsistence and emerging farmers.
There are transaction costs barriers to participation that can only be overcome
by institutional innovation. Research is therefore needed to identify policy
options that will stimulate the transition of smallholder farmers to become
commercial operators.
This study aims to propose ways to alleviate or
remove constraints that inhibit participation in agricultural markets.
1.3
BACKGROUND
1.3.1 Exclusion of smallholders from markets in South Africa
South Africa is classified as an upper middle-income country (World Bank,
1997). A number of studies conducted in South Africa, however, show high
levels of poverty in rural South Africa. According to the poverty report (1998)
just under 50% of the population (that is about 19 million people) live in the
poorest 40% of households and are thus classified as poor. It is striking to
4
University of Pretoria etd
note that a majority of the poor are located in predominantly rural provinces,
such as the Northern Province and Eastern Cape. These provinces have
poverty rates of about 70% and 60% respectively. In particular, poverty is
manifest in the former homeland areas where Africans are located. These
homelands emanated from the separate development policies initiated at the
beginning of the twentieth century. Under these policies the former homeland
areas were provided with inadequate infrastructure and services. Generally,
farmers in these areas had poor access to resources such as land, credit
facilities and technology (van Rooyen, 1995).
The process of exclusion must be seen in its historical context and has caught
several authors’ attention. Terreblanche (1998) and Vink and van Zyl (1998)
have provided an historical account of how the exclusion started in the
beginning of the century as white monopoly got established in both the
political and economic sectors. In the process, blacks were kept in the fringes
of socio-economic development. In agriculture, the unfairness of the system
came to light when the Tomlinson recommendations were tabled espousing
the improvement of conditions for blacks in the reserves (Houghton, 1956;
Kirsten, 1994), which, incidentally, were not accepted by the then government
(Anonymous, 1956; Kirsten, 1994). Instead it was viewed that the exclusion
of blacks, then referred to as indigenous people, from the markets had to do
with realistic physical conditions such as transport and climate, but mostly
depended on their unwillingness to integrate into a western economic system
(Anonymous 46, 1957). This resulted in different agricultural policies being
applied to white commercial agriculture and to black small-scale farmers in
the homelands (Vink, Kirsten and van Zyl, 1998) which changed land use
patterns and affected farm incomes for both groups of farmers(ibid, 1998;
D’Haese and Mdula, 1998).
Because of the restrictive setting in the homelands, households largely
depended for their income on jobs in areas reserved for whites.
Many
households attempted multiple coping strategies to provide for their livelihood
(Mekuria & Moletsane, 1996; May, 1998).
For example, members of a
household would be involved in subsistence farming on small plots, and at the
5
University of Pretoria etd
same time they would commute or migrate to and from the place of work.
Despite the effort put into it, these activities could not provide enough income
to move households out of poverty. In turn this had negative implications for
food security. In general, however, small-scale farming has always been the
mainstay of coping strategies in food security. Parallel to the poverty pattern
mentioned earlier, most of the vulnerable households are located in rural
provinces such as the Northern Province.
1.3.2 Smallholders can survive economically
Small-scale farmers have continued to produce in the face of unfavourable
conditions. In fact, one of the paradoxes of existence of smallholder farmers
pertains to their sustainability in spite of harsh circumstances. In South Africa,
the small-scale farmers have subsisted on “uneconomic farm units” as a result
of the Land Act of 1913 that excluded small-scale farmers from owning land.
Moreover, these farmers have had very limited support services, which made
it difficult for them to operate economically.
In line with Schultz’ hypothesis of small but efficient, several studies have
established that smallholder farmers do have comparative advantages in the
use of resources (Ngqangweni, 2000; McIntire and Delgado, 1985), implying
that they use resources efficiently, that is, resources are not wasted. The
study by Ngqangweni has shown what kind of activities smallholders in the
Eastern Cape could pursue profitably and with an acceptable level of
efficiency.
Some of the activities, such as indigenous beef and citrus
production, showed considerable potential and good opportunities under low
fixed-cost technologies and irrigation conditions.
Heavy infrastructure
investments boost the 'per unit costs' but better marketing arrangements
lower transaction costs, and this boosts the returns to the farmer and
consequently to society as a whole.
In general small-scale farmers have the advantage of flexible family labour
resources and can allocate labour to activities with higher marginal returns.
For example, dryland low-value farming is normally left for the old men and
6
University of Pretoria etd
women since able-bodied men and women are involved in migratory jobs, and
younger children go to school. When farming returns increase, for example
through access to more land, or irrigation, more household members are
involved and supplement this with hired labour.
In this way a number of
smallholder farmers make profits and are able to survive economically.
Pertinent characteristics of such farmers are access to market, information
and assets.
1.3.3 Smallholders survival creates linkages for economic growth
As mentioned before, smallholders dominate the former homeland rural
areas.
Where these farmers are active and successful, other non-farm
economic activities emanate as a result. Successful smallholders create a
demand for non-farm sector goods (retail). This is apparent in many rural
settings of South Africa. For example, a typical sample village will have a
retail store located close to arable lands. Small rural towns with more nonfarm business enterprises are located adjacent to thriving farming activities
(normally irrigated farming). So, the linkages with other sectors get stronger
when farming can generate more income, and this, in turn, is a direct result of
market participation.
In his study, Ngqangweni (2000) established that consumption- or demandside linkages which are derived from supported smallholder agriculture in the
Eastern Cape Province matched those recorded in studies of similar
situations in Africa and Asia (Delgado et al, 1998).
These linkages were
strengthened by cash inflow into the rural areas in the form of non-farm
incomes from urban areas. The study asserts that the relationship provides
opportunities for tradable smallholder agriculture to be a significant source of
the required initial income injection, which comes from sale of local
agricultural tradables. These arguments provide an additional motivation why
it is important to improve the productivity and increase the level of sales of
small-scale farmers. Through these increased levels of income as a result of
increased sales, farmers stimulate a range of non-farm activities in the
economy, which provide job opportunities for the rural poor.
However,
7
University of Pretoria etd
increasing on-farm productivity and increasing the level of sales are
hampered by several constraints.
1.3.4 There are transaction costs barriers that require new institutions
A range of constraints and barriers limits smallholder participation in the
agricultural market. As a result most of the smallholder products are wasted
after harvesting or sold at very low prices. Because of the uncertainty about
prices, many farmers would take any price offered by buyers when there is a
chance to participate. Farmers generally do not have the required information
and means to locate better markets. Many a time reliable markets are located
further away and are difficult to access. Only farmers with assets such as
vehicles are able to move around in search of a better market. When one
visits market centres, it is not uncommon to meet farmers who used their own
vehicles to get to the market. These farmers are also better informed about
various buyers and are normally well connected with neighbouring (white)
farmers. There is frequent road traffic between white farms and communal
villages where smallholder farms are located. This implies that farmers with
assets can interact more effectively; on the other hand it also means that the
majority of small-scale farmers are out of touch with these markets.
It is, thus, evident that a range of transaction cost barriers prevents smallscale farmers from participating in commercial markets. It follows that this
lack of commercial activity by small-scale farmers does, ultimately, not lead to
any of linkage benefits as anticipated. There is therefore a need for alternative
institutions that can overcome barriers to market participation. In other parts
of Africa (such as Ethiopia) farmers have the alternative to use brokers to
market their grain (Gabre-Madhin, 1999). In South Africa, however, some
farmers (particularly maize producers) tend to engage institutions such as cooperatives and millers to take their grain for processing, and storage, and
these sometimes provide transport services. Despite this a large proportion of
this grain is consumed and only a minor portion is sold, implying the
persistence of barriers to remunerative options.
8
University of Pretoria etd
Research is therefore needed to identify and suggest policies and strategies
to overcome transaction cost barriers. This is based on the argument that
transaction costs prevent market participation. Furthermore it is expected that
such research will show which the policy interventions most needed are.
1.4
HYPOTHESES
The main hypothesis of the study is that farmers facing lower transaction
costs will participate more in the agricultural markets than those farmers
facing high transaction costs. These transaction costs reflect the character of
the market, but are mainly embedded in household characteristics and their
economic environment.
As a consequence farmers respond to market
barriers by opting for alternative market institutions.
The specific hypotheses to be tested are the following:
Farmers with better access to information are likely to participate in the
market, other things the same. The more information farmers have about the
market, the more they will participate in the market:
•
Extension contact makes farmers aware of possible market outlets for
their products. As such, farmers with better contacts have a better
chance of participating in the markets. This doesn’t necessarily lead to
higher levels of participation.
•
Education allows farmers to interpret information about the market. So,
farmers with better education are more likely to participate in the
market. The opposite is also true; the lower the education level the
less the market participation.
•
Proximity to markets allows farmers to contact potential markets for
information about the market conditions. Even when farmers are busy
selling, proximity allows them to present the products to the market in
9
University of Pretoria etd
time at lower costs.
As such, the proximity to market centres is
negatively related to market participation.
•
Good road conditions to the markets make it possible for farmers to
market the products cheaply.
Those farmers facing good road
conditions tend to participate more in the markets than those who face
poor road conditions do.
Farmers with more assets (or increased wealth) are likely to participate more
in the market, other things the same:
•
The size of the farm (land) used for production is positively related to
market participation. When farmers have more land their production
will be higher, thus making it sufficient for market participation since the
per unit transaction costs will be lower due to the economies of scale.
The more the farmer can produce the more will be marketed.
•
Ownership of vehicles will increase market participation. This allows
farmers to access information about the market and be in a position to
deliver products.
•
Farmers owning more livestock will participate more in markets.
Livestock ownership tends to serve as a security for risk of market
failure on the one hand, and contributing to productive assets on the
other hand.
•
Access to liquid assets, such as non-farm and pension earnings allows
farmers to invest in marketing activities. Access to non-farm income
and pensions will lead to more market participation.
The higher the risk or uncertainty farmers face, the less likely they will
participate in agricultural markets. The risk attitude of farmers emanates from
the structure of the household:
10
University of Pretoria etd
•
Households headed by females are less likely to participate in the
market.
This leads to higher transaction costs since women are
regarded as lacking credibility as contractual parties owing to the
perception that courts (particularly tribal) will favour men in the event of
a dispute with a woman.
•
Age is positively associated with participation in agricultural market
since older farmers may be more experienced in marketing
management and tend to have stronger networks and more credibility,
thus facing lower transaction costs. This relationship is expected to be
stronger than the alternative hypothesis that younger farmers are less
risk averse.
•
The size of the household is negatively related to participating in the
market.
Normally, household members are both production and
consumption units. When there are fewer opportunities to contribute
productively, household units will be more of consumption unit, as is
the case in the developing areas of South Africa.
That is, larger
households have more mouths to feed and therefore less to sell.
1.5
ANALYTICAL METHODS
The study employs two analytical methods to test the above-mentioned
hypothesis:
Firstly, descriptive statistics is applied to the basic characteristics of the
sample households in order to assess the difference in the household
participation.
This employs both frequency and means to describe the
households.
Secondly, selectivity models are applied to identify and test significant factors
of market participation. The selectivity models involve two-step estimation
11
University of Pretoria etd
similar to the Heckman’s two-stage procedure.
Firstly, probit models are
estimated to determine the factors affecting decision to participate. Then,
heckits (OLS accounting for selectivity bias) are estimated in the second
stage to estimate the significant factors contributing to the level of
participation. The two-step selectivity procedure is similar to the tobit model
decomposing the probability to participate and the level of participation hence tobit models are also estimated to validate the selectivity models.
The analysis is based on the information collected in the Northern Province.
The subsequent sections overview the salient features of the Northern
Province from which the study sites were selected.
The procedures for
selecting sampled households will also be discussed, which is followed by a
discussion of the agricultural setting of the study area.
1.6
THE STUDY AREA
1.6.1 Overview of the Northern Province
1.6.1.1
Background of the province
The Northern Province is situated in the far northern part of South Africa,
which is, interestingly, situated in the far southern part of Africa. The Province
is adjacent to the Northwest Province, Gauteng and Mpumalanga and shares
borders with Botswana, Zimbabwe and Mozambique (Map 1). The Northern
Province covers 9,6 % of South Africa's total area, amounting to 116 824 km².
This subsection and subsequent subsections in the section are based on
NPDA (1996) and DBSA (1998).
The Province is divided into six regions: viz. Northern Region, Lowveld
Region, Central Region, Southern Region, Western Region and Bushveld.
The regions are further divided into sub-regions and/or districts.
12
N1
MESSINA
AN
AN
DZ
SOUTPANSBERG
M n in g in isi
M u le n zh e
Kh u b vi
M a ku ya
Sh o n g w a n e
TH
AB
AM
OO
PO
LOWVELD
REGION
NAPHUNO
1
NA
LE
TA
BA
PH
UN
O
PHALABORWA
2
2
SEKHUKHUNELAND
N O R TH ER N
P R O V IN C E
MHALA
Pietersburg
N e lsp ru it
N
G
NEBO
M a fike n g
A
G
M PU M A LA N G A
Pre to ria
Jo h a n n e sb u rg
M P U M A LA N G A
N O R TH W EST
FR EE STA TE
Hom o
H la n e ki
Va ltyn
M a p e la
Tsh ire n zh e n i
D o p e ni
Ph ip h id i
D zin d i
RITAVI 3
K
PAR
G a -M o th ib a
M a ja
G a -Th a b a
Le kg o th o a n e
M o g a no
Th a b a kg o n e
13
14
15
16
17
18
19
20
21
22
23
24
13
NAMAKGALE
SOUTHERN
REGION
N O RTH
W EST
D ith a b a n e n g
M a th a b a th a
Su c c e ss
D o o rn d ra a i
Ko p e rm yn
1
12
GIYANI
RITAVI 1
1
POTGIETERSRUS
21
RIT
AV
I2
10
98
6
4 7
5
3
2
NG
RO
KE
MO 3
WATERBERG
WARMBAD
1
2
3
4
5
6
7
8
9
10
11
12
BA
TA
LE
G2
ON
KER
MO
11
Pietersburg
Potgietersrus
Thohoyandou
BOLOBEDU
SESHEGO
15
19 18
14
CENTRAL
REGION
16
THABAZIMBI
NA
NI
U
DO 23
L
IONA
NAT
1
ELLISRAS
17
NI
WA
VU
POTGIETERSRUS
I
AN
AN
DZ
G
R
U
SB
R
TE
IE
P
G
ON
ER
OK
M
24
DZ
A
UM
CH
BO
WESTERN
REGION
AN
OY
OH
TH
20 22
TE
BOTSWANA
2 3 °S
MUTALE
GER
KRU
NORTHERN
REGION
Louis Trichardt
I
MOZAMBIQUE
SOUTPANSBERG
U
N o rth ern Pro vin ce
BUSHVELD
REGION
3 1 °E
ZIMBABWE
MAP 1. THE DISTRICTS AND SITES IN THE STUDY AREA
SO U TH A FR ICA
K W A Z U LU /
N A TA L
Kim b e rle y
Blo e m fo n te in
N1
N O R TH ER N C A P E
S
LE
O
T
HO
D u rb a n
N elspruit
G A U T EN G
EA STER N C A P E
Pretoria
Bish o
W ESTER N C A P E
C a p e To w n
Po rt Eliza b e th
13
University of Pretoria etd
The Northern Province can also be divided into several topographic zones. In
the east there is a flat to gently undulating Lowveld plain, at an altitude of 300 to
600 m, bounded in the west by the northern Drakensberg escarpment and
Soutpansberg, with steep slopes and peaks up to 2 000m above sea level. The
almost level Springbok flats in the south lie at an altitude of 900 m, while the
Waterberg and Blouberg to the north, with undulating to very steep terrain, reach
2 000 m. The north-west zone is a flat to undulating plain, which slopes down to
the north and west, at 800 to 1 000 m.
The province falls in the summer rainfall region. The Lowveld region is hot and
dry, with no frost and an average rainfall of less than 500 mm per annum. The
mountains are cooler and wetter, with rainfall of 1500 mm or more in places. To
the west, the rainfall varies from 600 mm on the Springbok flats to less than 400
mm on the Botswana border, where it can be extremely hot in summer. Dryland
cultivation can only be practised on the Springbok flats and on the eastern
escarpment and its foothills.
The major rivers are the Limpopo in the north, and the Olifants and Letaba
further south, all of which drain eastward.
The Limpopo only flows strongly
occasionally, while the Olifants and Letaba are heavily utilised for irrigation,
especially east of the escarpment. Most part of the province is very dry. Drought
is an ever-present threat in the north, and a growing human population places
considerable pressure on the existing resources especially in the Letaba
catchment area.
Black and red fertile clay soils occur on the Springbok flats, with reddish-brown
sandy loams to the north and west. The mountains have deeper, highly leached
red soils in the wetter areas, with exposed rock where the climate gets drier.
Reddish-brown, gravelly soils, which have low fertility, predominate in the
Lowveld, with the best agricultural soils being alluvial soils next to most of the
rivers. The Province does not have much high potential agricultural soil.
14
University of Pretoria etd
Of the estimated total of 12 million hectares, 67% (8 million ha) is utilised as
agricultural land. Of these 8 million hectares of farmland, nearly 10% (0.8 million
ha) is utilised as arable land, 67% (5.4 million ha) as natural grazing, 18.4% (1.5
million ha) for nature conservation, 1.1% (0.088 million ha) for forestry and 2%
(0.16 million ha) for other purposes. About 76% of arable land (0.61 million ha) is
allocated to dryland cultivation and forms the most important kind of cultivation
occurring in the Northern Province.
In 1995 about five million people resided in the Province, making it the fourth
largest province, with approximately 13% of the country's total population (DBSA,
1998). Of the five million people, 90.8% resided in rural areas, thus the rural
inhabitants of the Province made up 22% of the country's total rural population.
In 1995 the Province had an annual population growth rate of 3 to 4%, and a
population density of 38 persons per square kilometre. Of the 5.2 million people,
55% (2.86 million) were female. About 48% of the population were 15 years old
or less. The Province had the lowest human development index of the country,
namely 0.47.
1.6.1.2
The economic structure
The Northern Province contributed only 3.6% to the GGP in the RSA during
1994. The biggest contribution, 49% to the GGP in the Northern Province, during
this period came from the tertiary sector: trade, transport, finance, community
services, government and other producers. The primary sector, consisting of the
agricultural and mining sectors, was the second biggest contributor to the GGP
(with 32.7%). Of this, 32.7%, agriculture contributed with 9.6%.
Agriculture has grown by a mere 3.8% between 1980 and 1991.
The low
contribution to the GGP may be due to drought in the latter years of this period.
Despite this, the agricultural sector had a reasonable growth performance index
15
University of Pretoria etd
(GPI), resulting in the Province topping the rest of the country with respect to
GPI.
In 1993 the gross income from agriculture was R1414 million. Of this amount,
17.3% was from field crops, 46.2% million from horticultural products, and 37%
from livestock. These statistics show the importance of horticultural and livestock
products in the province.
The total expenditure on intermediate goods and services, i.e. those items used
in production, was R865 million. Of this amount farm feed, maintenance and
repairs, and fuel contributed with 22.3%, 15.8% and 15.7% respectively. When
expenses are subtracted from gross income, the difference is the contribution of
agriculture to the GDP (Gross Domestic Product) amounting to R558 million.
The real net income from agriculture in the Northern Province increased from
R156 million in 1983 to R252 million in 1988, but declined rapidly in the period
from 1991 to 1993 (NPDA, 1996). Thus, when other indirect farming expenses
such as indirect taxes and depreciation are subtracted from the contribution to
GDP, a net loss of R28 million from agriculture was realised in 1993.
The economically active population in the Northern Province was 1 215 508 in
1994 (DBSA, 1998). The contribution to employment by the agricultural sector in
this Province is 20.1%. According to the expanded definition of unemployment,
the Northern Province has at present an unemployment rate of 47%.
1.6.1.3
Farming enterprises
The main farming enterprises in the Northern Province focus on the production of
vegetables (NPDA, 1996).
Within the Province the production of vegetables
contributes an average of about 22% to the gross income from agriculture. On
average the production of vegetables contributed approximately 18% to the total
16
University of Pretoria etd
gross income from vegetable production in South Africa. From the composition
of horticultural products in the Northern Province, it is clear that the production of
vegetables is the most important (49.1%), followed by citrus fruit (25.9%) and
subtropical fruit (17.5%).
The production of nuts forms a small part of the gross income from agriculture in
the Northern Province, yet it makes up about 43% of the gross income from nuts
in South Africa. The production of citrus and subtropical fruit in the Northern
Province contributes about 64% to the gross income from citrus and subtropical
fruit in South Africa.
Animal products are the second largest generators of gross income from
agriculture within the Northern Province (NPDA, 1996).
The main animal
products produced in the Province are beef, poultry, fresh milk and dairy
products, and pork. From the composition of animal production, beef (53.6%) is
the most important, followed by poultry (22.2%), fresh milk and dairy products
(11.8%) and pork (7.9%).
The contribution of field crops to the gross income from agriculture in the
Northern Province is relatively small. On average, for 1983, 1988, 1991 and
1993, field crops contributed 22.7% to the gross agricultural income of the
Province. Cotton only contributed 5.9%, tobacco 4.7% and maize 4% to the
gross agricultural income.
The composition of field crops produced in this
Province is cotton (24.3%), tobacco (21.1%), maize (17.4%) and sunflower seed
(10.7%).
On average the production of cotton in the Northern Province
contributed more or less 56% to the total gross income of cotton produced in
South Africa, sisal 51%, cowpeas 38% and tobacco 22%. Important field crops
for smallholder sector are maize, grain sorghum and beans.
The Pietersburg Fresh Produce Market was declared a national fresh produce
market in May 1995. The activities of the market have generally been small,
17
University of Pretoria etd
compared with other national fresh produce markets in the country. As of the
end of 2000 only two market agents, namely the National Potato Association and
the Northern Transvaal Cooperative, operated on this market. There are at least
260 other formal marketing outlets in the Province. Seventeen of these are grain
and oil grinding mills, 13 processors, 12 abattoirs, some eight canners and
preservers, 24 suppliers and distributors, 154 meat markets, 19 dairies and 13
fruit and vegetable markets.
1.6.2 Selection of study sites
For the purpose of the research the Northern Province was divided into five
regions in 1997 when the survey was conducted: the Northern Region, Lowveld
Region, Central Region, Southern Region and the Western/Bushveld Region.
The reason for stratification is that the regions emerged from different
administrations (of the former Lebowa, Venda, Gazankulu and central RSA),
which provided different support services to farmers, and thus might lead to
different transaction costs. All five regions were selected. Within each region
one district was randomly selected by picking from a shuffled pile of district
names in the region. An average of five sites was to be selected from each
district, with an additional site. In two districts (in the Southern and Northern
Region) all six sites were visited. In other regions the five sites were regrouped
to as few as three sites (Map 1). Table 1.1 shows the distribution of sites and the
respondents by regions.
Table 1.1: Distribution of research sites and respondents
Region
Northern
Lowveld
Central
Southern
Western/ Bushveld
Total
No of sites
6
3
4
6
3
22
Respondents
24
18
29
57
29
157
Group discussions
4
2
4
2
3
15
18
University of Pretoria etd
The Northern Region consists mainly of districts of the former Venda homeland
areas (which include Dzanani, Malamulele, Mutale, Thohoyandou and Vuwani),
some patches of the former Gazankulu homeland (mainly Malamulele area),
former Lebowa (most of the Bochum area), and also former RSA areas such as
Messina and Soutpansberg.
The region has predominantly good agricultural
land due to relatively high rainfall. The main town, Thohoyandou, is located in
the Thohoyandou district. Accordingly, this district was randomly selected. Six
research sites were randomly selected from a list of ward names provided by the
extension service. From each ward a list of farmers was drawn, thus a total of
twenty-four respondents were selected and interviewed. The information from
household interviews was followed up by group discussions that elaborated on
pertinent issues raised during the face-to-face interview. Four group discussions
were conducted in the Northern Region.
The Lowveld region comprises mainly the former Gazankulu districts (such as
Giyani, Hlanganani, Lulekani and Ritavi), some districts of the former Lebowa
(mainly Naphuno, Bolobedu, Namakgale and Sekgosese), as well as the areas of
the former RSA (Letaba and Phalaborwa). The lowveld areas of the region are
mainly horticulture (fruit) production areas. For the purpose of the study, the
Giyani district was selected. The Giyani district produces a range of agricultural
products.
The district is also the locus of the regional offices of former
Gazankulu, located in the town of Giyani. The research sites selected were
Hlaneki, Homo, and Mninginisi. About 18 households were interviewed from
these three sites. Only two group discussions were conducted.
The Central region comprises predominantly the former Lebowa districts of
Mankweng, Sekgosese, Seshego, and Bochum) and the Pietersburg districts of
the former RSA. Pietersburg serves as the capital city of the Province. The
Mankweng district was selected randomly from other districts.
Although the
northern areas of the central region are livestock producing areas, by contrast
Mankweng district, lying south east of Pietersburg, is a predominantly maize
19
University of Pretoria etd
producing area.
The research sites included GaMamabolo, GaMolepo,
GaThaba, as well as Koppermyne (or GaMaja and GaMothiba).
About 29
respondents were interviewed in four sites.
The Southern Region comprises areas of the former Lebowa districts
Sekhukhune, Nebo, and Thabamoopo. This region, located south of Pietersburg
is mainly arable with relatively low livestock production. The Thabamoopo district
was selected for the survey. Two sites were selected, namely GaMathabatha
and Dithabaneng.
The former site is situated in a mountainous area with
sufficient water, while the latter is dry. The latter site is located close to
Lebowakgomo, the regional offices of the Southern region.
About 58
respondents were interviewed. Two focus groups were involved.
Finally, the Western region mainly comprises the former Lebowa districts of
Mokerong (or Mahwelereng), Zebediela and Phalala, as well as the former RSA
areas of Potgietersrus, Ellisras, Thabazimbi, Warmbaths, and Waterberg.
Mahwelereng is close to Potgietersrus, while Phalala is closer to Ellisras. The
sites were picked from both Mahwelereng and Phalala. The motive was that
there was still no clarity as to whether Phalala would fall under the then proposed
Bushveld region. The Bushveld is an area further west in the Province, with the
major towns Naboomspruit, Nylspruit, Warmbaths and Thabazimbi. It comprises
predominantly the former RSA areas.
The Western region is relatively dry,
although farmers focus on maize production in Mahwelereng, and livestock in the
Phalala area.
1.6.3 Agricultural setting of the study sites
Agricultural production in the Northern Province is diverse. This is reflected by
the diverse agricultural production systems in the study areas. A majority of
households (almost 70%) across the regions tend to focus on the production of
field crops, dominated by maize in area planted and level of production. A typical
20
University of Pretoria etd
sample household would plant 1.56 ha of maize, which covers 50% of the arable
area. Only 26% of the households produce maize under irrigation. This applies
to some households in Mathabatha area in the Southern Region as well as to
Mapela in the Western region. Maize in the Central Region, the Lowveld Region
and the Northern Region is grown under dryland conditions. These areas tend to
have reasonable level of rainfall.
Other field crops are also grown in the area, though not on such a wide scale as
maize. They include grain sorghum, millet, beans and watermelons grown under
dryland conditions. Generally these are grown as a mixed-cropped system. In
such cases, planting methods are mainly through broadcasting during ploughing
(which is normally done by contractors). Other field crops such as wheat are
grown in selected irrigation projects such as in Mathabatha and Mapela.
In some instances farmers would farm a plot in a project setting, while growing
other field crops on individual plots.
Monocropping or intercropping is
emphasised in project settings, while mixed cropping is practised mostly on
individual plots.
Government extension officers guide the projects by
recommending the best practices. Their advice centres on practices such as
planting time and application of recommended types of fertilisers and seeds. In
the projects fertilisers and seeds are normally bought from co-operatives, or
supplied directly by agents.
Farmers in the projects are expected to follow
certain production programs to harmonise the provision of inputs and service,
with output marketing.
Some households in the study area are also involved in horticultural production.
Less than 1% of households surveyed grow subtropical fruits such as bananas.
These households are mainly in the Lowveld region, in particular at the Homo
irrigation scheme. Vegetables are grown across the regions, though by relatively
few households. Most of vegetables grown in the study area are cabbages,
spinach, tomatoes, potatoes etc. Horticultural crops are grown under irrigation.
21
University of Pretoria etd
Livestock production is prevalent in the Northern Province. The incidence of
higher livestock production, however, increases towards the north-western part of
the Province. It follows that most of the livestock is found in the Western Region.
The livestock categories include large-stock (mainly cattle), small-stock (mainly
goats) and poultry. The livestock is kept in kraals at night and allowed to graze in
communal camps during the day. Generally livestock is kept as precautionary
assets, disposed off only when there is a need for cash. Local buyers provide a
major market for small-stock, while large-stock is sometimes auctioned.
1.7
THE SURVEY AND DATA
1.7.1 Sampling procedure
As mentioned before, the procedure for sampling was stratified by region. All five
regions were selected. Within each region, districts were selected randomly from
a shuffled pile of district names. Within the district, extension wards (composed
of villages) were also selected randomly from a shuffled pile.
Sampling of
households involved obtaining a sample frame of farmers from the extension
office. Households were then randomly picked from the list. Where the list was
not available before visiting the research site, farmers were convened, matched
with the extension officer’s register and randomly selected for interview.
heads of the households were interviewed.
The
In the absence of the head
(husband), the wife or the second member was interviewed.
The main
respondent would provide most of the information, but consulted with other
household members where necessary.
1.7.2 Data collection
22
University of Pretoria etd
The information was collected in the Northern Province in 1997 following twostages; face-to-face interviews with individual farmers, and group discussions
with focus groups of farmers.
The face-to-face interviews were conducted with 157 randomly selected farmers.
All respondents were requested to answer a set of structured questions. The
respondents were given the opportunity to consult with other household
members. The responses from the face-to-face interviews were reviewed, and
based on this pertinent issues were identified.
These issues were then presented and discussed during a follow-up group
discussion. The group consisted of a number of farmers who then elaborated on
the issues. The groups were composed of farmers in the area; those who were
interviewed and those who were not. They were convened through extension
officers.
1.7.3 Variables collected
The instrument was designed to collect a range of information. This included
information about household structure, consumption of food and non-food items,
factors of production (land, labour, capital, human resource, natural resources,
infrastructure, and management), as well as crop and livestock production.
Not all the information was usable for the study. For the purpose of the study the
following information was utilised:
•
Amounts of production sold at the market. This pertained to livestock,
horticulture crops, maize and other field crops.
•
Characteristics of the household regarding gender and age of the
household head, as well as the size of the household. Other information
23
University of Pretoria etd
collected pertained to access to income and assets. This included nonfarm income, pensions, arable land, and livestock as well as transport
equipment.
Information reflecting the farmer’s access to market
information was also collected in terms of average household education,
contact with extension service and proximity to the nearest town where the
markets are.
The conditions of the roads to the markets were also
determined.
1.8
CAVEATS
This study focuses on farmers in the Northern Province. These farmers are not
necessarily representative of the total population of South African farmers. As
such, generalisation of the results may not be possible without taking note of
limitations.
The study also focuses on transaction costs as they affect smallholder farmers’
decisions and level to participate in output markets. This focus might give the
impression that transaction costs are the sole factors of market participation,
while in fact they form part of a range of other factors affecting an entrepreneur.
1.9
ORGANISATION OF THE STUDY
The study is organised in six chapters.
The second chapter discusses the
literature review of smallholder market participation with respect to transaction
costs.
The third chapter presents a theoretical and empirical model (with
estimation procedure) for analysing the effect of transaction costs in smallholder
market participation. The descriptive characteristics of households in the study
area are then presented in chapter four. The results of the model are presented
in chapter five. Finally, the summary is presented and conclusions are drawn in
chapter six.
24